Heterologous Immune Responses of Serum IgG and Secretory IgA Against the Spike Protein of Endemic Coronaviruses During Severe COVID-19

Abstract

Defining immune correlates of disease severity is important to better understand the immunopathogenesis in COVID-19. Here we made use of a protein microarray platform to detect IgG- and IgA-reactive antibodies in sera and saliva respectively, and assess cross-reactivity between SARS-CoV-2 and endemic coronaviruses (eCoVs). IgG responses against the full protein of spike, but not the S1 subunit, were significantly higher in convalescent sera of patients with severe disease compared to mild disease and healthy controls. In addition, we detected reactivity of secretory IgA to eCoVs in saliva of patients with severe disease, not present in patients with moderate disease or seropositive healthy controls. These heterologous immune responses are in line with non-protective cross-reactivity, and support a potential role for immune imprinting in the pathogenesis of severe COVID-19.

Keywords: COVID-19; SARS-CoV-2; immune imprinting; seasonal coronaviruses; spike protein.

Figure 1

IgG responses against structural viral epitopes in convalescent sera distinguishes hospitalized patients with severe disease from mild/asymptomatic disease. (A) IgG antibody responses against SARS-CoV-2 recombinant spike and nucleocapsid structural proteins measured by protein microarray in convalescent sera from COVID-19 patients (red box), mild/asymptomatic disease controls (blue box) and unexposed healthy controls (gray box) Data are represented as the median titers (min and max value). (B) Receiver operating characteristic (ROC) curve analysis of protein microarray for discrimination between convalescent patients with severe COVID-19 requiring ICU care and those with mild/asymptomatic disease. (C) Vaccine-induced IgG antibody responses in seronegative healthy individuals vaccinated with spike mRNA vaccine (BNT162b2). SARS-CoV-2 refers to the original Wuhan-1 strain. ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 2

IgG responses against spike protein of eCoVs are induced in convalesced patients with severe COVID-19. (A) IgG antibody titers against recombinant trimeric spike proteins and monomeric S1 subunits of eCoV 229E measured by protein microarray in convalescent sera from COVID-19 patients (red box), mild/asymptomatic disease controls (blue box) and unexposed healthy controls (gray box). (B) Same as in A for eCoV NL63. (C) Same as in A for eCoV HKU1. (D) Same as in A for eCoV OC43. (E) Same as in A for SARS-CoV-1. (F) Same as in A for MERS-CoV. Data are represented as the median titers (min and max value). Significance (one-way ANOVA) *P < 0.05, **P < 0.01, ****P ≤ 0.0001.

Figure 3

IgG titers against eCoVs of the genus Betacoronavirus correlate in patients with a fatal disease outcome. (A) Correlations between serum IgG responses against SARS-CoV-2 and eCoVs of the genus Alphaviruses. (B) Correlations between serum IgG responses against SARS-CoV-2 and eCoVs of the genus Betaviruses. Spearman’s rank correlations are shown for each pair of spike trimer antigens, presented as R value and significance.

Figure 4

Positive IgA responses detected in saliva of patients with severe COVID-19. Heatmap visualization of IgA antibody titers against recombinant trimeric spike proteins and monomeric S1 subunits of SARS-CoV-2 and eCoVs measured by protein microarray in serially diluted saliva samples of COVID-19 patients with moderate and severe disease compared to seropositive healthy controls. Positive responses are indicated by red squares. ICU, intensive care unit; PU, pulmonary unit.